I am generating a 10Hz square wave with a CD4060. The output is fed into a BD139 that drives a potted HV module (input resistance 18.5K). Power supply is 5VDC. The purpose to to flash a neon lamp. I am aware there are other ways of doing this, but this is part of a larger project.

The generator works perfectly on its own. But when I connect the HV module, the DC level shifts up about a volt, and ringing appears across the top of the wave form. When I connect the neon lamp, the ringing increases about three fold. Please refer to photos provided. Ignore the speaker in the schematic.

Can someone please suggest a solution so I can understand the problem? It appears to be related loading, but the 18.5K seems reasonable enough to me. Maybe it is some kind of feedback from the bridge? Nothing I have tried so far removes it, even in the slightest.

 

I don't think what you are seeing can be described as ringing. That behavior occurs in an RLC circuit in response to a step function and the amplitude decays exponentially. What I think is going on is that you have a bunch of little antennas on your breadboard that are picking up high frequency noise from some source. A PC board would be a partial solution or you could redo the breadboard "Manhattan style" with leads as short as possible. Noise problems like this are difficult to track down but there is only one way to do it and that is locate the source.

https://www.google.com/search?q=Man...lAhUsn-AKHdYHDWEQsAR6BAgHEAE&biw=1600&bih=757
Ignore the images on the top line

 

I don't think what you are seeing can be described as ringing. That behavior occurs in an RLC circuit in response to a step function and the amplitude decays exponentially. What I think is going on is that you have a bunch of little antennas on your breadboard that are picking up high frequency noise from some source. A PC board would be a partial solution or you could redo the breadboard "Manhattan style" with leads as short as possible. Noise problems like this are difficult to track down but there is only one way to do it and that is locate the source.

https://www.google.com/search?q=Man...lAhUsn-AKHdYHDWEQsAR6BAgHEAE&biw=1600&bih=757
Ignore the images on the top line

@Sir Kit
Why do you think your HV module has an input "resistance" of 18.5K? What matters is the input impedance of the HV module, a very different thing than resistance and something you cannot measure with a DVM/VOM. Inside the HV module the input likely connects to many different components (resistors, capacitors, inductors, transistors, diodes, etc.). I suspect that what you are seeing is simply that the HV module input has a capacitor connected across its power input wires (a very common configuration). When your circuit pulls its output "high", the HV module capacitance charges and the module outputs some voltage. When your circuit essentially disconnects itself from the HV module, the capacitor in the module slowly discharges. By "disconnect" I refer to the use of an emitter follower; although the emitter can pull a load from ground toward the positive supply, it cannot pull a load to ground. Only the 220ohm resistor and the LED are pulling the HV module input (and its capacitance) toward ground. The LED becomes an open circuit when the voltage across it is less than 1-2 volts and discharge of the cap then ends. That happens to be about the same time that the next pulse occurs. The "ringing" you see at the top of the waveform is likely simply due to the varying load of the module as an oscillator/inverter in the module rapidly switches on and off.

 

Let me ask you this: Is the ringing a problem of just a cosmetic defect? With your setup ithe plastic plug-in breadboard) it might well be that Q1 is oscillating at a radio frequency. If it is not a problem for you, just ignore it (this should not be ignored for engineers designing a product that needs RFI qualification). If it is a problem try placing a 100 or 220 ohm resistor in series with the base of Q1 and see if it stops the oscillation.

 

If it is a problem try placing a 100 or 220 ohm resistor in series with the base of Q1 and see if it stops the oscillation.

Thank you for your suggestion, but adding the series resistance had no effect on the oscillation. Is there not something I can do to dampen it? I am looking for a relatively clean HV output.

 

@Sir Kit
Why do you think your HV module has an input "resistance" of 18.5K? What matters is the input impedance of the HV module, a very different thing than resistance and something you cannot measure with a DVM/VOM. Inside the HV module the input likely connects to many different components (resistors, capacitors, inductors, transistors, diodes, etc.). I suspect that what you are seeing is simply that the HV module input has a capacitor connected across its power input wires (a very common configuration). When your circuit pulls its output "high", the HV module capacitance charges and the module outputs some voltage. When your circuit essentially disconnects itself from the HV module, the capacitor in the module slowly discharges. By "disconnect" I refer to the use of an emitter follower; although the emitter can pull a load from ground toward the positive supply, it cannot pull a load to ground. Only the 220ohm resistor and the LED are pulling the HV module input (and its capacitance) toward ground. The LED becomes an open circuit when the voltage across it is less than 1-2 volts and discharge of the cap then ends. That happens to be about the same time that the next pulse occurs. The "ringing" you see at the top of the waveform is likely simply due to the varying load of the module as an oscillator/inverter in the module rapidly switches on and off.

As far as I know, the HV module is simply a Cockcroft-Walton bridge with no active components. Disconnecting the LED does not lessen the ringing. In terms of discharging any cap internal to the module, if that is the problem, would it help to install a bleeder resistor across the load (neon lamp)? Any further ideas would be appreciated.

 

I have since tried series and parallel caps in various positions, a bleeder resistor across the HV module input (killed it), and driving the BD139 with a Darlington stage.

I hate to give up. Is there really no way to eliminate or reduce this oscillation?

 

What is the frequency of the "ringing"?

 

I have since tried series and parallel caps in various positions, a bleeder resistor across the HV module input (killed it), and driving the BD139 with a Darlington stage.

I hate to give up. Is there really no way to eliminate or reduce this oscillation?

The best suggestion I can give would be try resolve whether the "ringing" you see is a spurious oscillation or something originating within the HV module. To that end I offer a simplified driver circuit, adding a PNP transistor as pull-down. You could try the circuit with and without R1, as per @DickCappels suggestion. You could also try with and without the 220 ohm R and the LED connected to the Q1/Q2 emitter junction. What does the HV module input drive look like with the complementary driver?

 

I don't see any power supply bypass capacitors in your schematic. I hope you have some in your circuit near the transistor.
Bypass capacitors, among other things, keep the power supply wiring from acting like a resonant circuit.

 

Based on the latest helpful suggestions, I disconnected the LED and added bypass caps. There was no change to the ringing.

Then I built the complementary pair driver as per teekay6's post. While it did not change the ringing, it did square up the wave form significantly. That's a keeper.

I have measured the ringing frequency at approx 10KHz. Its wave form resembles an ascending ramp with a 20% off time between cycles. Does that offer any clue or hope of a solution?.

 

I have measured the ringing frequency at approx 10KHz.

Is that perhaps the internal switching frequency of the HV module.

 

Thank you for your sharing, I have learnt so much from your passage.

 

Is that perhaps the internal switching frequency of the HV module.

I had assumed it was a simple CW bridge. But maybe not. There is a small trimpot to adjust the output voltage.

Here is a circuit diagram of what appears to be a similar unit. So I guess I will just have to accept a bit of switching noise.

 

Hello,

As far as I know the capacitors in a Cockroft-Walton generator have the same value.



https://en.wikipedia.org/wiki/Voltage_multiplierhttps://en.wikipedia.org/wiki/Cockcroft–Walton_generator
Bertus

 

I had assumed it was a simple CW bridge. But maybe not. There is a small trimpot to adjust the output voltage.

Here is a circuit diagram of what appears to be a similar unit. So I guess I will just have to accept a bit of switching noise.

@Sir Kit
I would not call the signal you see "noise" as it is essentially no different than power supply ripple due to a periodic load. It is not random. It seems to have a known source: the HV module. Unless it is promulgating to other parts of the circuitry (the bypass caps should help prevent that), I would consider it to be an expected signal of no consequence. It's possible that other drive configurations, more complex than emitter followers, would lessen the amplitude of the 10KHz signal, but why bother if it's not hurting anything?

 

@Sir Kit
I would not call the signal you see "noise" as it is essentially no different than power supply ripple due to a periodic load.

Yes, it is a continuous 10KHz ripple that varies in amplitude with load on the module.

I have no HV probe to look at it with, but the ripple may in fact be absent from the module's output due to the CW bridge. IOW it may appear only at the input as shown in my original post.